MEDIDA Y PAGO:
12. INSTALACIONES ELECTRICAS VOZ Y DATOS
12.3 SALIDAS PARA TOMACORRIENTES
A careful analysis of the behavior of occupational (industry) mobility over the business cycle is called for but is beyond the scope of this paper. Here we just point out some basic but important observations. We find aggregate occupational and industry mobility only mildly procyclical (almost acyclical).28 As is evident from Figures 3 and 4, however, this masks the
very different patterns of behavior of various age-education population subgroups over the business cycle. For the workers with education levels of high school or less, it is strongly countercyclical for workers younger than 30 and strongly procyclical for those older than 30. Occupational (industry) mobility of college educated workers in all age groups is little
28
This result is consistent with Loungani and Rogerson (1989) who find that (gross) labor reallocation across two-digit industries does not display any pronounced cyclical pattern.
affected by the business cycle conditions. These findings suggest a potentially important type of worker heterogeneity so far overlooked in the analysis of the welfare costs of business cycles.
6
Conclusion
The analysis in this paper was designed to provide a set of key facts characterizing the patterns of occupational and industry mobility in the US over the 1968-1993 period. We document that the level of occupational and industry mobility is high and has increased substantially over the period. In addition, we show that this is a profound change in the labor market that has affected a large fraction of the labor force. For instance, occupational mobility has increased for most age-education groups, and its rise was not driven by an increased flow of workers into or out of a particular one-digit occupation.
The high level of occupational mobility that we documented in this paper may seem surprising to an academic economist. It may be less surprising if a curious economist took the time to question other people about their careers. He or she might hear a story like this one: “Yes, I very much enjoy my position as a journalist writing for a newspaper chain. I used to be a high school English teacher,” or, “my career as a small business consultant grew from my love of training employees in my chain of donut shops. Before that I was a police officer. I started out as a truck driver, however.”29 More rigorously, we must reiterate
that the Retrospective Files available from the PSID represent the best data on annual occupational and industry mobility available in the United States. We thus have a lot of confidence in the levels and trends of mobility that we have reported.
We defined occupations and industries using the one, two-, and three-digit classifications
29
These examples are based on ”Making Career Sense of Labour Market Information” - the 1998 guide to Canadian Career Councilors. This Guide prepared by Canadian Career Development Foundation contains
utilized by the 1970 Census of Population and provided by the Panel Study of Income Dynamics for the 1968-1993 period. The examination of the occupational titles suggests that human capital is likely to be three-digit rather than one- or two-digit specific. A close look at the three-digit occupation classification reveals that skills accumulated in a given three-digit occupation may not be easily transferable to another three-digit occupation. For example, if an economics professor becomes a psychologist or a librarian, then, despite staying in the same one- and two-digit occupation, she would not be able to use most of her human capital accumulated while being in economics. Results in Kambourov and Manovskii (2002b) confirm this intuition. Specifically, they find that the returns to ten years of occupational experience are as high as 12.33% at the one-digit level, 15.19% at the two-digit level, and 19.00% at the three-digit level. Thus we suggest that researchers interested in calibrating their models using observations on career mobility should use the levels and trends in occupational mobility at a three-digit level documented in this paper.
Of course, most of the time workers who switch occupations tend to move into occupations that are relatively close to the occupation they have left. It remains an open research question, however, how one can develop a metric of how close various occupations are from each other in terms of skill transferability. It remains true, however, that everything else being constant, the average worker with ten years of occupational tenure would see his wages decline by at least 19% upon an occupation switch because many of the skills accumulated in the previous occupation are not used any longer and new skills need to be developed.
In view of the sharp rise in mobility documented in this paper, the next logical step is the investigation of the causes of its increase. Kambourov and Manovskii (2002a) suggest that the variability of occupational demand shocks has increased over time. They also argue in a general equilibrium model that the increase in mobility was not likely to be caused by a
decline in the costs of switching occupations. Other potential causes of the increased mobility include the usual suspects such as technological change, globalization and international trade, changes in government regulation and labor force unionization.
An intriguing research question is to relate changes in occupational mobility to changes in the growth rate of productivity. It may not be a coincidence that the increase in occupational mobility we have documented has coincided with a much discussed slowdown in productivity growth.
To conclude, with this paper we would like to bring the issue of occupational mobility to the attention of the profession. Why do people switch their occupations so often? How do people choose their occupations? Why has occupational mobility increased so much in the last 30 years? Is the increase in occupational mobility the missing link that would finally help us understand the changes in wage inequality and the aggregate performance of the economy? These and many other related questions beg economists’ attention. We believe that answering them will significantly advance our understanding of the labor markets.
References
Beaudry, P., and D. A. Green (2000): “Cohort Patterns in Canadian Earnings: As- sessing the Role of Skill Premia in Inequality Trends,” Canadian Journal of Economics, 33(4), 907–936.
Bernhardt, A., M. Morris, M. Handcock, and M. Scott (1999): “Trends in Job Instability and Wages for Young Adult Men,” Journal of Labor Economics, 17(4), 65–90. Bertola, G.,andA. Ichino(1995): “Wage Inequality and Unemployment: United States vs. Europe,” in NBER Macroeconomics Annual, ed. by B. Bernanke, and J. Rotemberg. The MIT Press.
Blanchard, O. J., and P. Diamond(1990): “The Cyclical Behavior of the Gross Flows of U.S. Workers,” Brookings Papers on Economic Activity, 2, 85–143.
Felli, L., and C. Harris (2003): “Firm-Specific Training,” mimeo, London School of Economics.
Gottschalk, P., and R. Moffitt (1994): “The Growth of Earnings Instability in the U.S. Labor Market,” Brookings Papers on Economic Activity, 2, 217–272.
Hagedorn, M., G. Kambourov, and I. Manovskii (2004): “Worker Mobility in the United States and Germany: a Primer,” mimeo, University of Pennsylvania.
Jovanovic, B. (1979): “Job Matching and the Theory of Turnover,” Journal of Political Economy, 87(5), 972–990.
Jovanovic, B., and R. Moffitt (1990): “An Estimate of a Sectoral Model of Labor Mobility,” Journal of Political Economy, 98(4), 827–852.
Jovanovic, B., and Y. Nyarko (1997): “Stepping Stone Mobility,” Carnegie-Rochester Conference Series on Public Policy, 46(1), 289–326.
Jovanovic, B., and P. Rousseau (2003): “Specific Capital and the Division of Rents,” mimeo, University of Chicago.
Kambourov, G., and I. Manovskii(2002a): “Occupational Mobility and Wage Inequal- ity,” mimeo, The University of Western Ontario.
(2002b): “Occupational Specificity of Human Capital,” mimeo, The University of Western Ontario.
(2002c): “Rising Occupational and Industry Mobility in the United States: 1968- 1993,” mimeo, The University of Western Ontario.
(2004): “A Cautionary Note on Using (March) CPS Data to Study Worker Mobil- ity,” mimeo, The University of Pennsylvania.
Khun, P. (2003): “Effects of Population Aging on Labor Market Flows in Canada: Ana- lytical Issues and Research Priorities,” mimeo, University of California, Santa Barbara. Krusell, P., L. E. Ohanian, J.-V. R´ıos-Rull, andG. L. Violante(2000): “Capital-
Skill Complementarity and Inequality: A Macroeconomic Analysis,” Econometrica, 68, 1029–1053.
Ljungqvist, L., and T. J. Sargent (1998): “The European Unemployment Dilemma,” Journal of Political Economy, 106(3), 514–550.
(2002): “The European Employment Experience,” Discussion Paper 3543, Centre for Economic Policy Research.
Loungani, P., and R. Rogerson (1989): “Cyclical Fluctuations and Sectoral Realloca- tion: Evidence from the PSID,” Journal of Monetary Economics, 23(2), 259–273.
Lucas, R. J., and E. Prescott(1974): “Equilibrium Search and Unemployment,” Jour- nal of Economic Theory, 7, 188–209.
MaCurdy, T., and T. Mroz (1995): “Measuring Macroeconomic Shifts in Wages from Cohort Specifications,” mimeo, Stanford University.
Markey, J. P., and W. Parks II (1989): “Occupational Change: Pursuing a Different Kind of Work,” Monthly Labor Review, 112(7), 3–12.
Mathiowetz, N. A.(1992): “Errors in Reports of Occupations,” Public Opinion Quarterly, 56(3), 352–355.
McCall, B. P. (1990): “Occupational Matching: A Test of Sorts,” Journal of Political Economy, 98(1), 45–69.
Miller, R. A. (1984): “Job Matching and Occupational Choice,” Journal of Political Economy, 92(6), 1086–1120.
Moscarini, G., and F. Vella(2003): “Aggregate Worker Reallocation and Occupational Mobility in the United States: 1976-2000,” mimeo, Yale University.
Murphy, K. M., and R. H. Topel (1987): “The Evolution of Unemployment in the United States:1968-1985,” in NBER Macroeconomics Annual, ed. by S. Fischer. The MIT Press.
Neal, D. (1995): “Industry-Specific Human Capital: Evidence from Displaced Workers,” Journal of Labor Economics, 13(4), 653–677.
(1999): “The Complexity of Job Mobility Among Young Men,” Journal of Labor Economics, 17(2), 237–261.
Parent, D. (2000): “Industry-Specific Capital and the Wage Profile: Evidence from the National Longitudinal Survey of Youth and the Panel Study of Income Dynamics,” Journal of Labor Economics, 18(2), 306–323.
Parrado, E., and E. Wolff (1999): “Occupational and Industry Mobility in the United States, 1969-1992,” Working paper, C.V. Starr Center, New York University.
PSID (1999): A Panel Study of Income Dynamics: 1968-1980 Retrospective Occupation- Industry Files Documentation (Release 1) Survey Research Center, Institute for Social Research, The University of Michigan, Ann Arbor, Michigan.
Rosenfeld, C. (1979): “Occupational Mobility During 1977,” Monthly Labor Review, 102(12), 44–48.
Table 1: Coding Error Statistics.
Variable χ2(9) P rob > χ2 Average Increase
in Mobility 1-Digit Occupations 27.61 0.0011 0.1061 2-Digit Occupations 40.24 0.0000 0.1279 3-Digit Occupations 74.74 0.0000 0.2430 1-Digit Industries 32.06 0.0002 0.0949 2-Digit Industries 57.88 0.0000 0.1515 3-Digit Industries 64.30 0.0000 0.2505
1-Digit Ind.-Occ. Cells 18.80 0.0270 0.0143
2-Digit Ind.-Occ. Cells 24.28 0.0039 0.0275
3-Digit Ind.-Occ. Cells 47.64 0.0000 0.1085
Note. - The first column reports the Wald test statistic for the null hypothesis that the coding error is the same across all age-education groups while the second column reports the probability of not being able to reject the null hypothesis. The last column reports the average increase in occupational, industry, or occupation-industry mobility after 1981 due to the estimated coding error.
Table 2: Estimated Coefficient on the Time Trend in Mobility on the Overall Sample, Various Population Structures.
Population Structure
Variable Actual Average 1970 1980 1990 1-Digit Occupations 0. 0021 0. 0025 0. 0024 0. 0026 0. 0026 (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) 2-Digit Occupations 0. 0021 0. 0025 0. 0024 0. 0028 0. 0025 (0.0003) (0.0003) (0.0004) (0.0004) (0.0003) 3-Digit Occupations 0. 0012 0. 0018 0. 0015 0. 0022 0. 0016 (0.0004) (0.0004) (0.0004) (0.0004) (0.0004) 1-Digit Industries 0. 0018 0. 0021 0. 0023 0. 0023 0. 0021 (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) 2-Digit Industries 0. 0017 0. 0021 0. 0020 0. 0023 0. 0021 (0.0004) (0.0003) (0.0004) (0.0004) (0.0003) 3-Digit Industries 0. 0008 0. 0012 0. 0012 0. 0014 0. 0011 (0.0004) (0.0004) (0.0005) (0.0004) (0.0004) 1-Digit Ind.-Occ. Cells 0. 0014 0. 0016 0. 0016 0. 0016 0. 0017
(0.0002) (0.0002) (0.0002) (0.0002) (0.0002) 2-Digit Ind.-Occ. Cells 0. 0019 0. 0021 0. 0019 0. 0023 0. 0022
(0.0002) (0.0002) (0.0003) (0.0002) (0.0002) 3-Digit Ind.-Occ. Cells 0. 0010 0. 0015 0. 0015 0. 0018 0. 0014
(0.0003) (0.0003) (0.0004) (0.0003) (0.0003)
Note. - Each cell represents estimates of the time trend in mobility on the overall sample for various population structures. The observed mobility is corrected for the estimated coding error after 1980. The estimates are obtained from an OLS regression of the corresponding mobility variable on a constant and a time trend for the 1969-1993 period. Standard errors are in parentheses.
Table 3: Average Estimated Switch Probability for Various Age-Education Groups and Different Occupational, Industry, and Industry-Occupation Classifications.
Occupation Industry Industry-Occupation Cells 1-Digit 2-Digit 3-Digit 1-Digit 2-Digit 3-Digit 1-Digit 2-Digit 3-Digit Variable (1) (2) (3) (4) (5) (6) (7) (8) (9) Group 11 0.2644 0.3040 0.3881 0.2130 0.2634 0.2695 0.1537 0.1985 0.2444 Group 12 0.2193 0.2619 0.3198 0.1702 0.1919 0.2153 0.1117 0.1480 0.1825 Group 21 0.1766 0.2032 0.2405 0.1108 0.1253 0.1286 0.0890 0.1086 0.1213 Group 22 0.1350 0.1597 0.1986 0.1095 0.1218 0.1373 0.0671 0.0911 0.1211 Group 31 0.1333 0.1513 0.1736 0.0736 0.0857 0.1016 0.0427 0.0649 0.0853 Group 32 0.0752 0.0810 0.1064 0.0634 0.0739 0.0759 0.0397 0.0426 0.0503 Group 41 0.0826 0.0997 0.1192 0.0798 0.0872 0.0807 0.0587 0.0612 0.0598 Group 42 0.0702 0.0691 0.0722 0.0436 0.0467 0.0583 0.0163 0.0240 0.0230 Group 51 0.0707 0.0763 0.0951 0.0399 0.0482 0.0502 0.0216 0.0261 0.0326 Group 52 0.0613 0.0684 0.0736 0.0297 0.0409 0.0474 0.0249 0.0420 0.0373
Note.- Each cell represents the average (over the 1969-1993 period) predicted switch probability for various age-education groups. The predictions are from a probit regression. The binary dependent variable indicates whether there was a switch on a one-, two-, or three-digit level, respectively. The sample was divided into 10 age-education groups ij, where i denotes the age group while j denotes the education group. By age individuals are divided into the following groups: 23-28, 29-34, 35-40, 41-46, 47-61. By education, individuals are divided into those who have 12 years of education or less and those who have more than 12 years of education. The independent variables in the regression include dummy variables dumij indicating the group
that the individual belongs to, time trend variables tdumij for each of the ten groups, structural break
variables breakij for each of the ten groups capturing the change in the coding methodology in 1980, and
Table 4: Estimated Time Trend Coefficients for Various Age-Education Groups. Different Occupation, Industry, and Industry-Occupation Cell Classifications.
Occupation Industry Industry-Occupation Cells 1-Digit 2-Digit 3-Digit 1-Digit 2-Digit 3-Digit 1-Digit 2-Digit 3-Digit Variable (1) (2) (3) (4) (5) (6) (7) (8) (9) Group 11 0. 0038 0. 0046 0. 0053 0. 0071 0. 0083 0. 0058 0. 0046 0. 0065 0. 0067 (0.0010) (0.0011) (0.0013) (0.0012) (0.0011) (0.0012) (0.0009) (0.0011) (0.0011) Group 12 0. 0037 0. 0040 0. 0055 0. 0023 0. 0027 0. 0026 -0. 0003 0. 0020 0. 0023 (0.0012) (0.0012) (0.0012) (0.0014) (0.0015) (0.0015) (0.0011) (0.0011) (0.0012) Group 21 0. 0031 0. 0048 0. 0026 -0. 0002 -0. 0001 -0. 0023 0. 0010 0. 0020 -0. 0003 (0.0010) (0.0012) (0.0012) (0.0008) (0.0008) (0.0008) (0.0006) (0.0008) (0.0007) Group 22 0. 0039 0. 0037 0. 0045 0. 0041 0. 0037 0. 0036 0. 0018 0. 0029 0. 0047 (0.0010) (0.0010) (0.0009) (0.0006) (0.0007) (0.0008) (0.0006) (0.0007) (0.0007) Group 31 0. 0030 0. 0030 0. 0013 0. 0005 0. 0005 -0. 0002 0. 0004 0. 0011 0. 0004 (0.0008) (0.0007) (0.0009) (0.0007) (0.0006) (0.0010) (0.0007) (0.0006) (0.0007) Group 32 0. 0020 0. 0013 -0. 0005 0. 0023 0. 0020 0. 0009 0. 0023 0. 0019 -0. 0002 (0.0010) (0.0010) (0.0009) (0.0009) (0.0009) (0.0009) (0.0004) (0.0005) (0.0009) Group 41 -0. 0003 -0. 0005 -0. 0023 0. 0033 0. 0026 0. 0002 0. 0038 0. 0022 0. 0004 (0.0014) (0.0014) (0.0015) (0.0007) (0.0007) (0.0009) (0.0006) (0.0008) (0.0011) Group 42 0. 0033 0. 0014 -0. 0006 0. 0004 -0. 0001 -0. 0005 0. 0009 0. 0010 0. 0001 (0.0013) (0.0010) (0.0012) (0.0009) (0.0012) (0.0008) (0.0005) (0.0006) (0.0008) Group 51 0. 0004 -0. 0001 -0. 0009 -0. 0002 -0. 0006 -0. 0012 -0. 0006 -0. 0010 -0. 0016 (0.0004) (0.0004) (0.0006) (0.0005) (0.0006) (0.0008) (0.0004) (0.0004) (0.0005) Group 52 0. 0011 0. 0013 -0. 0001 0. 0009 0. 0013 0. 0010 0. 0015 0. 0030 0. 0010 (0.0006) (0.0007) (0.0007) (0.0007) (0.0006) (0.0005) (0.0005) (0.0006) (0.0006)
Note.- Each cell represents the time trend in mobility for a specific age-education group over the 1969-1993 period. Standard errors are in parentheses. The sample was divided into 10 age-education groups ij, where i is the corresponding age group while j is the corresponding education group. By age individuals are divided into the following groups: 23-28, 29-34, 35-40, 41-46, 47-61. By education, individuals are divided into those who have 12 years of education or less and those who have more than 12 years of education. For each group ij, the observed mobility is corrected for the estimated coding error after 1980. Then the reported estimates are obtained from an OLS regression of the mobility for group ij on a constant and a time trend.
Table 5: Fraction of Workers Returning to Their Occupation or Industry, 1969-1980.
Fraction of Workers Returning After Variable One Year Two Years Three Years 1-Digit Occupation 0.1953 0.1073 0.0630 (0.0100) (0.0078) (0.0061) 2-Digit Occupation 0.1669 0.0870 0.0521 (0.0095) (0.0071) (0.0056) 3-Digit Occupation 0.1172 0.0567 0.0346 (0.0082) (0.0058) (0.0046) 1-Digit Industry 0.1861 0.0612 0.0608 (0.0098) (0.0058) (0.0061) 2-Digit Industry 0.1453 0.0477 0.0507 (0.0088) (0.0053) (0.0055) 3-Digit Industry 0.1188 0.0360 0.0476 (0.0081) (0.0047) (0.0053)
Note. - Each cell represents the fractions of workers who return to their occupation (industry) one year, two years, or three years after they have switched them. Standard errors are in parentheses. The results are obtained using the Retrospective Files for the period 1969-1980. For each of the years from 1970 till 1977 we identify those workers who have just switched their occupation (industry), and then we follow them for three years in order to determine the fraction that returns to their previous occupation (industry). The reported statistics are averaged over this period.
Table 6: Mobility Across Broad Occupational Groups.
A. Average Mobility Over the 1970-1975 Period
To Relative From 1 2 3 4 5 6 Size 1 91. 73 3. 70 2. 17 1. 31 0. 60 0. 49 17.45 (0.17) (0.11) (0.08) (0.06) (0.05) (0.05) (0.10) 2 4. 66 84. 05 5. 93 3. 13 1. 71 0. 54 13.50 (0.16) (0.23) (0.18) (0.13) (0.10) (0.05) (0.09) 3 3. 75 10. 53 78. 97 2. 64 4. 35 1. 82 11.03 (0.14) (0.23) (0.33) (0.13) (0.16) (0.09) (0.07) 4 1. 08 1. 85 2. 32 80. 78 10. 97 3. 01 27.78 (0.05) (0.07) (0.08) (0.17) (0.13) (0.08) (0.13) 5 0. 97 0. 95 1. 99 14. 18 74. 31 7. 60 21.41 (0.06) (0.05) (0.08) (0.18) (0.25) (0.16) (0.10) 6 0. 70 2. 43 1. 89 10. 30 17. 09 67. 59 8.60 (0.07) (0.12) (0.11) (0.24) (0.33) (0.38) (0.07) B. Average Mobility Over the 1982-1987 Period
To Relative From 1 2 3 4 5 6 Size 1 81. 50 9. 94 2. 78 4. 40 0. 46 0. 92 17.86 (0.22) (0.17) (0.10) (0.12) (0.04) (0.05) (0.09) 2 7. 81 76. 91 7. 82 5. 10 1. 33 1. 03 17.97 (0.15) (0.27) (0.17) (0.14) (0.08) (0.06) (0.10) 3 4. 59 14. 26 68. 01 4. 51 4. 40 4. 23 10.76 (0.16) (0.25) (0.31) (0.15) (0.16) (0.16) (0.08) 4 3. 46 4. 36 1. 69 77. 84 8. 33 4. 33 25.31 (0.09) (0.10) (0.06) (0.18) (0.13) (0.09) (0.11) 5 0. 79 1. 83 2. 24 11. 66 76. 60 6. 87 19.49 (0.05) (0.07) (0.09) (0.20) (0.26) (0.15) (0.10) 6 2. 16 4. 85 4. 99 9. 66 14. 67 63. 67 8.60 (0.11) (0.19) (0.19) (0.26) (0.28) (0.40) (0.06) C. Average Mobility Over the 1988-1993 Period
To Relative From 1 2 3 4 5 6 Size 1 80. 21 9. 94 3. 79 3. 56 1. 37 1. 12 19.24 (0.23) (0.18) (0.09) (0.11) (0.04) (0.06) (0.08) 2 8. 89 75. 15 7. 77 5. 37 1. 34 1. 48 19.67 (0.14) (0.25) (0.16) (0.12) (0.08) (0.06) (0.10) 3 5. 86 13. 44 68. 74 3. 19 4. 00 4. 77 11.80 (0.15) (0.21) (0.34) (0.17) (0.14) (0.15) (0.08) 4 3. 97 4. 92 1. 80 76. 63 8. 60 4. 07 22.44 (0.09) (0.09) (0.06) (0.22) (0.15) (0.09) (0.10) 5 2. 02 1. 64 1. 91 10. 43 76. 97 7. 04 18.26 (0.05) (0.07) (0.09) (0.19) (0.25) (0.15) (0.10) 6 2. 07 4. 08 6. 19 10. 55 15. 01 62. 10 8.89 (0.12) (0.16) (0.19) (0.23) (0.26) (0.34) (0.07)
Note. - Cell ij represents the average (over the period) percent of those working in occupation i in a given year who will work in occupation j the following year. Occupational groups are defined as: 1. Professional, technical, and kindred workers; 2.